Vladimir Ristich and Siyuan Liang contributed equally to this work.
Tolerization of dendritic cells by HLA-G
Article first published online: 15 MAR 2005
Copyright © 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
European Journal of Immunology
Volume 35, Issue 4, pages 1133–1142, April 2005
How to Cite
Ristich, V., Liang, S., Zhang, W., Wu, J. and Horuzsko, A. (2005), Tolerization of dendritic cells by HLA-G. Eur. J. Immunol., 35: 1133–1142. doi: 10.1002/eji.200425741
- Issue published online: 22 MAR 2005
- Article first published online: 15 MAR 2005
- Manuscript Accepted: 9 FEB 2005
- Manuscript Revised: 26 JAN 2005
- Manuscript Received: 11 OCT 2004
- Dendritic cells;
- Inhibitory receptors;
- Allograft survival
The expression of HLA-G at the fetal-maternal interface during pregnancy and in transplanted tissue makes this a key molecule in the acceptance of a semiallogeneic fetus and allogeneic transplant. Dendritic cells (DC) play a critical role in the control of innate and adaptive immune responses. DC are present in maternal decidua, but must be kept under tight control. Here we describe the mechanism of tolerization of DC by HLA-G through inhibitory receptor interactions. The HLA-G-ILT (immunoglobulin-like transcript) interaction leads to development of tolerogenic DC with the induction of anergic and immunosuppressive T cells. Using human monocyte-derived DC and ILT4-transgenic mice, we show that (i) HLA-G induces the development of tolerogenic DC with arrest maturation/activation of myeloid DC, (ii) HLA-G-modified DC induce differentiation of anergic and immunosuppressive CD4+ and CD8+ effector T cells, and (iii) the gene expression profile provides evidence that HLA-G induces tolerogenic DC by disruption of the MHC class II presentation pathway. Ligation of ILT4 receptor on DC from transgenic mice diminished peptide presentation by MHC class II molecules and significantly prolonged allograft survival. These findings provide support that HLA-G is an important tolerogenic molecule on DC for the acceptance of a semiallogeneic fetus and transplanted tissue/organ.
HLA G is a human nonclassical MHC molecule with potential tolerogenic properties. The expression of HLA-G on the extravillous cytotrophoblast during pregnancy and its other unique functions suggest a special role for HLA-G in protection of the semiallogeneic fetus from recognition by the maternal immune system 1–3. In addition, the expression of HLA-G by allografts, by infiltrating mononuclear cells within transplanted tissues, by tumor and tumor-infiltrating antigen-presenting cells (APC) 4–6 propose immunosuppressive functions for HLA-G in the regulation of immune responsiveness. The major function of classical MHC class I molecules is the presentation of various peptides to specific CD8+ T cells, but HLA-G is limited in this respect 7–10. Another particular function of classical MHC class I molecules is to serve as a ligand for a variety of immune inhibitory and activating receptors expressed by immunocompetent cells. These receptor-ligand interactions control the activation and function of immunocompetent cells. HLA-G is a ligand for a number of immune inhibitory and activating receptors expressed by different cell types. Immunoglobulin-like transcripts (ILT, also known as LILR, and CD85) represent an immunoglobulin superfamily of inhibitory and activating receptors that are expressed in a different type of dendritic cells (DC). ILT2 and ILT4 receptors, the most characterized immune inhibitory receptors, are expressed predominantly on myeloid DC 11–13. ILT3, ILT7 are mainly expressed on plasmacytoid DC 14, 15. The expression of ILT on DC is tightly controlled by inflammatory stimuli, cytokines, and growth factors, and is down-regulated following DC activation 15. Recently, it has been shown that the presence of HLA-G on APC induces immunosuppressive CD4+ T cells 16. The mechanism of these findings still remains to be elucidated. We hypothesized that tolerogenic DC can be induced by selectively targeting inhibitory receptors on DC by their natural ligand, HLA-G. Here, we demonstrate that HLA-G induces tolerogenic DC, which are characterized by low-level expression of MHC class II and costimulatory molecules on the cell surface. These DC reduce the potential to stimulate allogenic T cells. In contrast to untreated DC, the maturation signal provided by allogenic T cells does not affect the level of expression of MHC class II and costimulatory molecules on tolerogenic HLA-G-treated DC, suggesting that HLA-G inhibits activation/maturation of DC. Moreover, tolerogenic DC induce expansion of alloreactive T cells, which are anergic and immunosuppressive. Analysis of the gene expression profile of tolerogenic DC shows that treatment of DC with HLA-G affects the cluster of genes that control antigen loading, processing, and presentation of MHC class II-peptide complex to CD4-positive T lymphocytes. Therefore, when the ILT4 receptor has been ligated by HLA-G in transgenic mice expressing ILT4 exclusively on DC, a significant decrease of peptide presentation to MHC class II-restricted CD4+ T cells has been shown. In addition, treatment of ILT4-transgenic mice with HLA-G significantly prolongs skin allograft survival. These findings show that HLA-G is an important tolerogenic molecule for DC and thus can be used for the induction of tolerogenic DC via interactions with ILT.
Cytokines and growth factors regulate the expression of ILT2 and ILT4 receptors on human monocyte-derived DC
The expression of ILT2 and ILT4 receptors is highly regulated by histone acetylation, which contributes to strictly controlled gene expression exclusively in the myeloid lineage of cells 17. We analyzed the ability of several cytokines and growth factors to enhance the expression of ILTs on DC. As shown in Fig. 1, monocyte-derived DC generated by the standard method, using GM-CSF and IL-4, have a very low level of cell surface expression of ILT2 and ILT4 receptors. Since it was previously shown that IL-4 reduces the expression of immune inhibitory receptors on murine B cells 18, we tested if IL-4 affects expression of the inhibitory receptors on human monocyte-derived DC. Removal of IL-4 from the protocol for the generation of DC increases the number of ILT2-positive cells from 1.9±0.24% to 46.7±3.39%, and ILT4-positive from 4.6±0.32% to 24.8±2.29% (Fig. 1B). There is growing evidence that up-regulation of ILT4 can be induced by antigen-specific T suppressor cells and certain cytokines 19, 20. In particular, it has been shown that IL-10 induces up-regulation of the inhibitory receptor ILT4 in monocytes 21. We determined that DC generated in the presence of GM-CSF and exposed to additional IL-10 for 48 h moderately increased the number of cells expressing ILT4 receptor (Fig. 1C). Moreover, a marked increase of ILT inhibitory receptors on human monocyte-derived DC was observed when the cells were treated with TGF-β (Fig. 1D, 72.6±6.43% of cells became ILT2 positive, and 42.5±3.41% of cells became ILT4 positive). Thus, we were able to determine optimal conditions for expression of the inhibitory receptors on human monocyte-derived DC generated in vitro, and we have used this protocol throughout the current studies. These data reflect in vivo situation, where expression of ILT is regulated by several soluble factors, including GM-CSF, TGF-β, and IL-10, which are present both in the placenta and in accepted allografts.
HLA-G-modified DC induce anergic and immunoregulatory CD4+ and CD8+ T cells
Previously we showed that the treatment of human monocyte-derived DC with HLA-G tetrameric complexes generates tolerogenic DC, with a decrease in the level of expression of MHC class II and costimulatory molecules 22. The status of tolerogenic DC can be reversed by several signals, including CD40L signal derived from allogeneic T cells, making the DC immunogenic and initiating T cell responses 23. As seen in Fig. 2, unmodified DC showed high level expression of MHC class II (HLA-DR) and the costimulatory molecules CD80 and CD86. These findings suggest that the maturation signal (CD40L) from allogeneic T cells activated/maturated unmodified DC. In contrast, the expression of HLA-DR, CD80 and CD86 in HLA-G-modified DC was reduced (Fig. 2). Altered expression of CD80 was observed mainly among the CD11c-high population of cells. Based on our findings, we conclude that HLA-G-modified DC are still resistant to the natural CD40L maturation signal provided by allogeneic T cells and does not change the expression of cell surface molecules that hallmark features of immature/tolerogenic DC. In general, the T cell interactions with tolerogenic DC lead to induction of various tolerogenic mechanisms via abortive expansion, deletion, and anergy of the antigen-specific T cells or the expansion/induction of regulatory T cells. Thus, we investigated the tolerogenic mechanism induced by HLA-G-modified DC on activation and differentiation of naive CD4+ and CD8+ T cells. As shown in Fig. 3A, the population of activated CD4+25+ T cells was strikingly decreased in the presence of HLA-G-modified DC. Additional analysis of proliferative CD4+25+ T cells revealed that HLA-G-modified DC raised the number of CTLA-4 (CD152)-positive cells measured by intracellular staining (Fig. 3B). Further analysis showed that in the presence of HLA-G-modified DC, allogeneic CD4+ T cells prominently reduce the capacity to produce IL-2 and IFN-γ (Fig. 3C, D). In addition, the number of CD4+ T cells with the capacity to produce IL-10 increased from 6.9%±0.52% to 16.4±2.21% (Fig. 3E). Our results provide evidence that HLA-G-modified tolerogenic DC negatively affect the expansion of highly activated CD4+ T cells following allogeneic stimulation and may facilitate the generation of a population of CD4+ T cells with regulatory properties. Analysis of CD8+ T cells revealed no difference in the production of IFN-γ and IL-2 by CD8+ T cells generated with HLA-G-treated or nontreated DC (data not shown). However, we determined the increase in number of CD8+CD28– T cells in the population that were generated by HLA-G-modified DC (Fig. 4A). In addition, these cells have an increased capacity to produce IL-10, as measured by intracellular staining (Fig. 4B). Thus, we conclude that HLA-G-modified DC can induce or expand the population of CD8+CD28– T cells with immunoregulatory properties.
Next we analyzed the proliferative capacity of CD4+ T cells activated by HLA-G-modified and unmodified DC. As shown in Fig. 5A, CD4+ T cells activated by HLA-G-modified DC showed significant reduction of spontaneous proliferation. Furthermore, the proliferative activity induced by anti-CD3 antibody has been significantly diminished in these cells. An additional signal provided by exogenous IL-2 augments the proliferative responses of activated CD4+ T cells through the IL-2 receptor. We determined that exogenous IL-2 greatly increased the proliferative capacity of CD4+ T cells activated with unmodified DC 8.6–12.4-fold increase (Fig. 5A). The potential of proliferative responses of CD4+ T cells generated with HLA-G-modified DC increased by 7.7–10.9-fold. These data indicated that the majority of CD4+ T cells activated by tolerogenic HLA-G-modified DC are in an anergic state. The development of anergy in these CD4+ T cells has been supported by our findings that the cells have a markedly reduced capacity to produce IL-2 as identified by intracellular staining. However, we also assessed the development of the CD4+ T cell population with regulatory T cell function by tolerogenic HLA-G-modified DC. The phenotype of these cells was characterized as CD4+CD25+CTLA-4+, with an increased capacity to produce IL-10 and decreased capacity for IL-2 production. Additional functional analysis revealed that the presence of these cells in the population of CD4+ T cells generated by HLA-G-modified DC inhibited the allogeneic proliferative response of naive CD4+ T cells (Fig. 5B).
HLA-G down-regulates genes controlling MHC class II assembly and presentation on human DC
To understand the mechanisms of tolerization of DC by HLA-G/ILT interactions, we analyzed the gene expression profile of human monocyte-derived DC exposed to HLA-G tetrameric complexes for 10, 18, and 24 h. cDNA probes were synthesized from total RNA samples and hybridized to gene-specific cDNA fragments spotted onto GEArray membranes. This GEArray includes 155 genes encoding proteins important for DC activation and maturation, such as cytokines, chemokines and their receptors, other cell-surface receptors, and signal transduction molecules. Genes for proteins involved in antigen uptake, processing, and presentation are also represented in this array. The comparison between HLA-G-treated and non-treated DC was conducted to identify differentially expressed genes affected by HLA-G. Expression levels of most of the transcripts were similar. However, treatment of DC with HLA-G specifically affected 11 genes, the majority of which were down-regulated (Table 1 and Fig. 6). The most affected genes were in the group involved in remodeling of specialized antigen-processing compartments and in MHC class II-restricted antigen presentation. These genes, which included the invariant chain (Ii, CD74) and genes whose expressed proteins play a critical role in peptide loading of MHC class II, showed a 2.8–3.4-fold decrease in transcript levels. HLA-DMB, HLA-DMA gene expression decreased 2.9-3.7-fold and 2.2-2.6-fold, respectively. Expression of IFN-γ-inducible lysosomal thiol reductase (IFI30, GILT) decreased 2.3–2.8-fold, and the DC-LAMP3 gene was down-regulated 2.1–2.7-fold. Transcription of the dynactin gene (Dynactin 2), which regulates the endoplasmic reticulum (ER)-Golgi intracellular trafficking of proteins, showed a 2.6–3.5-fold decrease, when cells were treated with HLA-G. The beta-glucan receptor, Dectin-1, a coreceptor for T cell activation, also showed decreased expression (2.1–2.6-fold reduction). A novel DC-specific multimembrane-spanning protein, DCSTAMP, was down-regulated 2.7–3.2-fold decrease by HLA-G. The function of this molecule on DC is still unknown. On the other hand, the expression levels of a very limited number of genes were up-regulated in the HLA-G-modified DC. The most frequently increased transcript was identified as cytokine IL-1β (2.7–3.2-fold increase). In some experiments, minor up-regulation of the CD83 gene was observed. Our results provide evidence that the major effect of the HLA-G interaction with DC expressing ILT4 and ILT2 receptors is altered expression of genes involved in MHC class II presentation.
|Accession no.||Gene name||Hybridization intensity||Fold decrease over non-treated|
|X00497||CD74||14407 (1901)||4873 (625)||3.0|
|X62744||HLA-DMA||1273 (131)||531 (70)||2.4|
|U15085||HLA-DMB||1368 (179)||416 (86)||3.3|
|BC021136||IFI30||4338 (634)||1741 (232)||2.5|
|BC000718||Dynactin 2||1006 (125)||348 (64)||2.9|
|AF313469||Dectin-1||29053 (3265)||13522 (1426)||2.2|
|AF305068||DC-STAMP||5315 (734)||1881 (229)||2.8|
|AB013924||DC-LAMP3||3223 (562)||1453 (259)||2.2|
|BC009470||PRKRA||5672 (593)||2650 (237)||2.1|
|BC020848||RNASE6||1609 (138)||632 (86)||2.5|
|M32599||GAPDH||64042 (7423)||63628 (5840)||1.0|
|M15330||IL-1β||2187 (267)||6778 (1046)||–3.1|
|S53354||CD83||1699 (317)||3396 (564)||–2.0|
HLA-G/ILT4 interaction diminishes MHC class II presentation in ILT4-transgenic mice
To test if the HLA-G/ILT4 interaction affects the ability of MHC class II molecules to present peptides to T cells, we used an antigen-presentation assay with murine cells that express ILT4 receptor exclusively on DC. ILT4-transgenic mice were generated as described in Materials and methods section. ILT4-transgenic mice were born at the expected rate, bred well, and were phenotypically normal. Histological analysis of thymus, spleen, liver, brain, kidney, stomach, intestine and skin from 6-week-old mice revealed no major abnormalities. Phenotyping of the immune system by flow cytometry demonstrated that the major lymphoid populations CD4+ and CD8+ T cells, and B cells were present at the expected frequency and numbers in the thymus, spleen, lymph nodes and bone marrow of ILT4-transgenic mice. DC were also present in normal numbers and frequency in the thymus, bone marrow, spleen. We assessed the expression of ILT4 receptor in transgenic mice using monoclonal antibodies specific for this receptor. ILT4 was constitutively expressed on all CD11c+ cells in spleen, bone marrow, and thymus (Fig. 7A, B and data not shown). Bone-marrow-derived DC and splenic DC from ILT4-transgenic and nontransgenic control mice were exposed to HLA-G tetramer and then assayed for their ability to present a range of concentrations of the murine MHC class II-binding peptide, chicken ovalbumin OVA323–339, to the T cell hybridoma DO11.10, which is specific for this peptide bound to I-Ab. No difference in peptide presentation has been determined between the untreated DC from the ILT4-transgenic and control C57BL/6 mice. Moreover, as shown in Fig. 7C, the exposure of DC to HLA-G from control mice have an insignificant affect on antigen presentation of peptide by MHC class II molecules. An additional maturation signal provided by LPS significantly increased the presentation of peptide by DC from nontransgenic control (Fig. 7D). In contrast, a decrease in peptide presentation was seen when DC from ILT4-transgenic mice have been exposed to HLA-G (Fig. 7C). Even less efficiency to present MHC class II-bound peptide to the specific CD4+ T cells was assessed in LPS-treated DC from ILT4-transgenic mice after exposure to HLA-G (Fig. 7D). These results provide strong evidence that the HLA-G/ILT4 interaction on DC targeting MHC class II compartments leads to a marked decrease in the efficiency of antigen presentation on MHC class II.
HLA-G/ILT4 interaction on recipient DC significantly prolongs skin allograft survival
To investigate the role of the HLA-G/ILT4 interaction in vivo on modulation of DC function in recognition of alloantigens, we used a skin allograft model. C57BL/6 (B6) mice very strongly recognize the MHC class II-disparant mutant mouse, B6.CH-2bm12 (bm12) carrying the I-Abm12 alloantigen 24. Skin grafts from bm12 to B6 were rejected at 9.5 days in control B6 mice, which received no treatment (Fig. 8A), whereas skin graft survival was prolonged in B6 mice, which had received HLA-G tetramers (MST=14.5, n=8, p<0.05). This prolongation is probably the result of interaction of HLA-G with murine paired immunoglobulin-like receptor, PIR-B 25. A significant prolongation of allograft occurred in ILT4-transgenic mice that received HLA-G tetramer (MST=42.4, n=10, p=0.0003), 75% of the grafts survived for >40 days. Furthermore, CD4+ and CD8+ T cells from spleen of ILT4-transgenic mice targeted with HLA-G tetramer less efficiently produced IL-2, as measured by intracellular staining, compared to CD4+ and CD8+ T cells from B6 mice that received HLA-G tetramer (Fig. 8B). Our results demonstrate that targeting of recipient DC with HLA-G/ILT4 can down-regulate the immune response of allogeneic T cells in vivo and prolong skin allograft survival in nonimmunosuppressed recipients.
We initiated the present study to evaluate the role of HLA-G in the induction of tolerogenic DC and to identify the mechanistic basis of this process. HLA-G is a unique molecule expressed mainly during pregnancy on the surface of cytotrophoblast cells, which invade the maternal decidua and play an important role in establishing maternal-fetal tolerance. The maternal decidua is represented by several types of immunocompetent cells including DC. DC are highly potent immunostimulatory cells for the efficient priming of T cell responses. In addition, DC play a pivotal role in the induction of antigen-specific unresponsiveness or tolerance. The control of DC activation /maturation is an important factor for establishing tolerance. DC may actively be rendered tolerogenic by a number of mechanisms, including the naturally-occurring CD8+CD28– regulatory T cells, which up-regulate inhibitory receptors on DC and disrupt CD40-induced B7–1 and B7–2 expression 21. HLA-G is a natural and specific ligand for ILT2 and ILT4 inhibitory receptors and might play an essential role in the control of DC activation/maturation, leading to the development of tolerogenic DC. The ligands for the ILT include several MHC class I and nonclassical molecules. Shiroshi et al. 26 demonstrated that ILT2 and ILT4 bound to HLA-G with a three-to-fourfold higher affinity than other classical MHC class I molecules. These findings support our hypothesis that the HLA-G/ILT interaction may play a significant role in the regulation of DC.
Here we demonstrate the mechanisms of the induction of tolerogenic DC by HLA-G. This mechanism involves inhibition of maturation and differentiation of myeloid DC. Myeloid DC exist in two functionally distinct states, immature (iDC), and mature. Monocytes cultured in the presence of GM-CSF differentiate into iDC. An additional maturation signal, provided by CD40L from allogeneic activated T cells, transforms iDC into mature DC. We determined that iDC that highly express ILT2 and ILT4 receptors when treated with HLA-G and stimulated with allogeneic T cells still maintain a stable immature-like phenotype (CD80low, CD86low, HLA-DRlow) with the potential to induce anergy in allogeneic T cells. In addition, the HLA-G interaction with DC that highly express ILT2 and ILT4 receptors resulted in down-regulation of several genes involved in the MHC class II presentation pathway. A lysosomal thiol reductase, IFN-γ-inducible lysosomal thiol reductase (GILT), abundantly expressed by professional APC, was significantly reduced. The repertoire of primed CD4+ T cells can be influenced by DC expression of GILT, as in vivo T cell responses to select antigens were reduced in animals lacking GILT after targeted gene disruption 27. The HLA-G/ILT interaction on DC interferes with the assembly and transport of MHC class II molecules to the cell surface, which might result in less efficient presentation or expression of structurally abnormal MHC class II molecules. Loading of exogenous peptides onto MHC class II molecules has been shown to be critically dependent on H2-M (HLA-DM, human). Cells from mice with a targeted mutation in the H2-M gene are unable to present intact proteins and have a markedly reduced capacity to present exogenous peptides (10-to 20-fold reduction) 28. We determined that HLA-G markedly decreased the transcription of invariant chain (CD74), HLA-DMA, and HLA-DMB genes on human monocyte-derived DC highly expressing ILT inhibitory receptors. Analysis of transgenic mice expressing the ILT4 receptor revealed that the HLA-G/ILT4 interaction also affects peptide presentation to the specific CD4+ T cells. In vivo experiments showed that targeting of DC by HLA-G/ILT4 interactions significantly prolong allograft survival from donor skin with mismatched MHC class II antigen. In this model the indirect pathway of allorecognition was the major target. The prolongation of graft survival was associated with down-regulation of T cell activation and development of T cell anergy. However, the stimulation of T cell responses via the direct pathway of allorecognition is considered more sufficient to mediate an allograft rejection response and was part of the mechanism of allograft rejection. One of the approaches to target the direct pathway of allorecognition involves the induction of donor-specific tolerance. The HLA-G-modified tolerogenic DC can be used for induction of donor-specific tolerance in a nonimmunosuppressed recipient.
Interaction of HLA-G with DC involves a variety of immune inhibitory and activating receptors. ILT4 is the major inhibitory receptor on myeloid DC. Our data supports the hypothesis that HLA-G/ILT4 interaction plays a dominant role in the induction of tolerogenic DC. Further analysis will determine the role of HLA-G/ILT2 interaction in modulation DC functions.
Based on our findings, we conclude that tolerogenic HLA-G-modified DC induce a population of CD4+ T cells that are anergic. However, the phenotyping analysis of T cells revealed an increase in a fraction of cells that are CD4+CD25+CTLA-4+ and CD8+CD28– with a high capacity of IL-10 production. The phenotype of these cells and functional data support for special immunosuppressive/regulatory properties of these cells. The suppressive/regulatory properties of these cells and their utility in the induction/maintenance of hyporesponsiveness or tolerance induced by the HLA-G/ILT interaction warrant further investigation.
Materials and methods
Generation of human monocyte-derived DC and T cell isolation
To generate DC, CD14+ monocytes were cultured in RPMI 1640 medium, supplemented with 10% FCS, 2 mM L-glutamine, 10 mM Hepes, 1 mM sodium pyruvate, 100 U/ml penicillin, and 100 μg/ml streptomycin (complete medium) in the presence of 50 ng/ml GM-CSF (R&D Systems, Minneapolis, MN) for 5 days. For some experiments, IL-4 (R&D Systems) was added. DC were also treated with 40 ng/ml of IL-10 (e-Bioscience, San Diego, CA), or 20 ng/ml of TGF-β (e-Bioscience) during final 48 h of culture. The resulting DC were washed and cultured with HLA-G tetramer or without tetramer. HLA-G1 tetrameric complexes have been generated as described previously 7. For isolated CD4+ and CD8+ T cell subsets, elutriator-purified lymphocytes were subjected to immunomagnetic negative selection (Miltenyi Biotec, Auburn, CA).
T cell activation of cells via TCR/CD3 complex was assessed by stimulation of 2×105 purified CD4+ T cells with graded concentrations of plate-bound anti-CD3 mAb and 50 ng/ml of anti-CD28 mAb (e-Bioscience) in complete medium for 72 h. During the last 16 h of culture, 1 μCi of [3H]thymidine was added, and cellular incorporation was determined. For inhibition of allogeneic proliferation, graded numbers of T cells generated after culturing with unmodified DC or HLA-G-modified DC were cocultured with allogeneic MLR for 5 days and proliferation of cells was measured by 16 h thymidine incorporation.
Antibodies and flow cytometry
Human DC were stained with mAb anti-CD11c-APC/S-HCL-3, anti-CD123-PE/9F5, anti-CD80-PE/2D10.4, anti-CD86-FITC/IT2.2, anti-CD83-FITC/HB15e, anti-HLA-DR-FITC/LN3, anti-CD85j-FITC/GHI/75, or purified anti-ILT4/42D1 followed by FITC-labeled goat anti-rat IgG. Human T cells were stained with mAb anti-CD4-APC/RPA-T4, anti-CD8-FITC/RPA-T8, anti-CD28-PE/CD28.2, anti-CD25-APC/BC96, or anti-CD152-PE/14D3. Murine DC were stained with mAb anti-CD11c-APC/HL3, anti-CD11b-FITC/M1/70, anti-MHC class II-PE/M5/114.15.2, or purified anti-ILT4/42D1 followed by FITC-labeled goat anti-rat IgG. All primary and secondary reagents were purchased from BD-PharMingen (San Diego, CA) or from e-Bioscience. Cells were incubated with primary antibodies for 30 min at 4°C. When necessary, conjugated secondary antibodies were used. A BD Biosciences FACSCalibur (Mountain View, CA) was used for data acquisition and CellQuest software was used for analysis.
Intracellular staining of cytokines
Human lymphocytes (1×106) were stimulated with PMA (50 ng/ml) and ionomycin (500 ng/ml) for 5 h with the presence of BD GolgiPlugTM (BD Biosciences). Cells were harvested and stained with APC-conjugated anti-CD4 mAb and FITC-conjugated anti-CD25 mAb or FITC-conjugated anti-CD8 mAb and APC-conjugated anti-CD25 mAb. Permeabilized cells were, respectively stained with PE-conjugated anti-CD152 mAb, or PE-conjugated anti-IL-2 mAb, or PE-conjugated anti-IL-10 mAb, or PE-conjugated anti-IFN-γ mAb. Murine cells were stained with FITC-conjugated anti-CD4 mAb and APC-conjugated anti-CD8 mAb. Permeabilized cells were stained with PE-conjugated anti-IL-2 mAb. Cells were analyzed using FACSCalibur with CellQuest Analysis software.
RNA isolation and DNA microarray analysis
Total RNA from monocyte-derived DC treated with HLA-G beads was isolated with RNA STAT-60TM Kits (Tel-Test, Friendswood, TX). Monocyte-derived DC without treatment was used as a control and processed accordingly. Five micrograms of total RNA was used for cDNA probe synthesis. cDNA probe was generated with GEArray TrueLabelling-RT kit (SuperArray Bioscience, Frederick, MD). Following denaturation, the cDNA probe was hybridized to Human Dendritic & Antigen Presenting Cell Gene Array at 60ºC for 16 h. Data were analyzed by the Image Quant 1.2 Software (Amersham Biosciences, Piscataway, NJ) with STORM™ 840 Gel and Blot Imaging System (Amersham Biosciences). Signal for each transcript was normalized by comparing to the housekeeping gene GAPDH.
Generation of transgenic mice expressing ILT4 on dendritic cells
ILT4 cDNA (1.8 kb fragment) was subcloned into the EcoRI site of an expression vector containing the β-globin expression cassette and regulatory elements controlling the expression of CD11c in mice (provided by Dr. T. Brocker, Institute for Immunology, Munich, Germany). After the nucleotide sequence of the ILT4 expression vector was confirmed, the 8.3 kb plasmid-free fragment was microinjected into the pronuclei of fertilized oocytes of C57BL/6 mice. The expression of ILT4 receptor on murine DC has been determined by flow cytometry using an anti-ILT4 mAb. Two independent lines of ILT4-transgenic mice have been established. Experiments were conducted in accordance with institutional guidelines for animal care and use.
Peptide presentation assay
I-Ab restricted DO11.10 T hybridoma cells (2×105) were plated together with 2×105 DC and titrated amounts of specific OVA327–339 peptide. After 24 h, 80 μl supernatant from each well was added to 5×105 CTLL-2 cells to indicate IL-2 production. The responses of the peptide-specific hybridoma were assayed by IL-2 production.
Specific pathogen-free C57BL/6 (H-2b) mice and ILT4-transgenic mice (H-2b) (8–10 weeks of age) were used as skin graft recipients throughout the study. Recipient mice received HLA-G-coupled microspheres 25. Microspheres (5x106) were injected intravenously 7 days before skin grafting. Donor skin was from MHC class II-disparate B6.CH-2bm12 (bm12, H-2b) mice. Allogeneic skin grafts were performed by standard methods as described 25. All skin grafting survival data were analyzed using the Student's t-test.
The authors are very grateful to Drs. Marina Cella and Marco Colonna for providing reagents and critically reviewing the manuscript. We also thank Dr. Leszek Ignatowicz for help with the peptide presentation analysis. We especially thank Dr. Rhea-Beth Markowitz for helpful discussions and critical reading of the manuscript. Research support was provided by NIH grant RO1 AI 055923 to A. Horuzsko.